Teams of mobile robots often need to divide up subtasks efficiently. In spatial domains, a key criterion for doing so may depend on distances between robots and the subtasks' locations. This research considers a specific such criterion, namely how to assign interchangeable robots to a set of target locations such that the makespan (time for all robots to reach their target locations) is minimized while also preventing collisions among robots. We provide an overview of a scalable multiagent dynamic role assignment system known as SCRAM (Scalable Collision-avoiding Role Assignment with Minimal-makespan). SCRAM uses a graph theoretic approach to map agents to target locations such that our objectives for both minimizing the makespan and avoiding agent collisions are met. SCRAM scales to thousands of agents as role assignment algorithms run in polynomial time.

PDF, PS, HTML

In Proc. of 13th Int. Conf. on Autonomous Agents and Multiagent Systems (AAMAS), May 2014. Accompanying videos at http://www.cs.utexas.edu/~AustinVilla/sim/3dsimulation/AustinVilla3DSimulationFiles/2013/html/scram.html.

@inproceedings{AAMAS14-MacAlpine, title={{SCRAM}: Scalable Collision-avoiding Role Assignment with Minimal-makespan for Formational Positioning}, author={Patrick MacAlpine and Eric Price and Peter Stone}, booktitle={Proc. of 13th Int. Conf. on Autonomous Agents and Multiagent Systems (AAMAS)}, month={May}, note={Accompanying videos at <a href="http://www.cs.utexas.edu/~AustinVilla/sim/3dsimulation/AustinVilla3DSimulationFiles/2013/html/scram.html">http://www.cs.utexas.edu/~AustinVilla/sim/3dsimulation/AustinVilla3DSimulationFiles/2013/html/scram.html</a>}, url="http://www.cs.utexas.edu/users/ai-lab?macalpine::aamas14", year={2014} }